A dynamically reconfigurable stereoscopic/panoramic vision mobile robot head controlled from a virtual environment

We have built a mobile robotic platform that features an Active Robotic Head (ARH) with two high-resolution cameras that can be switched during robot operation between two configurations that produce respectively panoramic and stereoscopic images. Image disparity is used for improving the quality of the texture. The robot head switches dynamically, based on robot operation between the stereoscopic configuration and the panoramic configuration.

On-line extraction of stable visual landmarks for a mobile robot with stereo vision

This paper proposes a method to extract on-line stable visual landmarks from sensory data obtained by stereo vision so as to adapt to changes of environment. Given a two-dimensional obstacle map, the robot first extracts vertical line segments which are distinct and inside planar surfaces not near boundary edges as they are expected to be observed reliably from various viewpoints. However, the extracted feature information such as position and length include uncertainty due to errors of vision and motion. The robot then reduces the uncertainty by matching the planar surface containing the features to the map. These processes are performed on-line in order to adapt to actual changes of lighting and the scene depending on the robot's view. Experimental results in real scenes show the validity of the proposed method.     

嵌入式系统在机器人视觉中的应用

近年来,基于机器视觉的智能机器人在工业等各个领域中得到了广泛的应用。本文针对应用中遇到的一些问题,提出了将机器视觉与嵌入式系统相结合的思想。开发了一种通用的嵌入式系统平台,进行操作系统的移植和图像匹配等算法的研究,并将其应用于移动机器人的视觉导航,取得了较好的效果。     

Practical solution using a new approach to robot vision

A system has been devised for development and testing in robot vision applications. The hardware and software of the system are described. Keyboard manufacturing is taken as a typical example of robot vision applications. For this example, an account is given of the four main stages in a development and implementation programme: defining the requirements of the application; matching these to the vision system; developing the application software; integrating and installing the system. Further applications of robot vision are suggested.     

拟人机器人视觉系统的软件设计

清华大学“985”重点科研项目——拟人机器人技术和应用系统研究开发的拟人机器人TBIPR-I包括视觉系统，系统新的软件部分应用了Direct Show技术，采用了组件的形式实现了主要部分。将新旧系统进行了对比，并给出了部分代码实例。     

Perception of safe robot idle time in virtual reality and real industrial environments

The main objective of this study was to investigate human perception of safe idle time of an industrial robot in a virtual reality environment. Studying operators' perception of robot operational characteristics such as safe robot idle time can help develop hazard prevention strategies, and ultimately improve robot safety. Results of data collected from 32 participants showed that robot size and speed had significant effects on the perception of safe robot idle time. This study also examined operators' perceived acceptability level for the robot speed, perceived level of hazard of robot motion, perceived chance of error, and self-reported mental workload. Results of this study were compared to the findings of Rahimi, M., Karwowski, W. [1990. Human perception of robot safe speed and idle time. Behaviour & Information Technology 9(5), 381–389], in which their experiment was conducted in a real industrial environment. This study demonstrated the feasibility of testing human perception of dynamic moving objects in a virtual reality environment. The virtual reality technology is believed to be capable of modeling a complex machinery system such as a robotic system.Relevance to industry: Human perception of the operational characteristics of industrial robots is an important concern for robot safety since misperception can cause robot operators to err, which in turn can cause injuries and fatalities. Through this study we (1) understand human perception, safety behavior, and decision making in a robotic system and (2) demonstrate the capability of modeling a complex machinery system using virtual reality technology. Our experiments designed to study human perception of safe robot idle time could lead to safety interventions and guidelines or hazard prevention strategy development.     

基于单目视觉的移动机器人导航算法研究进展

基于单目视觉的移动机器人导航的研究,涵盖了机器视觉、模式识别和多目标跟踪多个领域．其算法框架不仅成功应用于移动机器人导航,还为目标检测、识别与跟踪领域的研究提供了可供参考的模型．该综述将以算法发展历史为脉络,结合一些典型系统,通过对关键技术和算法结构的分析比较,总结算法本身的发展前景和由此发展起来的可供相关研究参考的算法框架．     

Decomposition of transformation matrices for robot vision

The relationship between the three-dimensional coordinates of a point and the corresponding two-dimensional coordinates of its image, as seen by a camera, can be expressed in terms of a 3 by 4 matrix using the homogeneous coordinate system. This matrix is known more generally as the transformation matrix and it is well known that such a matrix can be determined experimentally by measuring the image coordinates of six or more points in space, whose three-dimensional coordinates are known.Such a transformation matrix can be derived analytically from knowledge of the camera position, orientation, focal length and scaling and translation parameters in the image plane. However, the inverse problem of computing the camera location and orientation from the transformation matrix involves solution of simultaneous nonlinear equations in several variables and is considered difficult.In this paper we present a new and simple analytical technique that accomplishes this inversion rather easily. This technique is quite powerful and has applications to a wide variety of problems in Computer Vision for both static and dynamic scenes. The technique has been implemented as a C program running under Unix and works well on real data.     

一种主动式全景立体视觉传感器设计

现有的三维激光扫描仪无法实时、全自动地计算空间三维信息,需要对获得的点云数据进行配准.为解决这些问题,研制了一种新型主动式全景立体视觉传感器( ASODVS),并且从影响ASODVS测距精度的各个因素进行了分析.实验结果表明:设计的ASODVS能自动、实时地进行空间物点深度测量,并得到相应的点云数据.     

机器人足球视觉系统中的实时图像处理

视觉系统是整个机器人足球系统的重要组成部分。根据机器人足球视觉系统的特点,提出基于游程长度编码(RLE)的实时快速图像处理算法。算法使用RLE对图像进行压缩,并且在处理阶段高效识别出图像中目标的尺寸和位置。压缩阶段算法的时间复杂度与图像尺寸成线性关系,图像处理阶段算法的时间复杂度与图像中目标的个数和每个目标所占的扫描行数成线性关系。     

Building facade detection,segmentation, and parameter estimation for mobile robot stereo vision

Building facade detection is an important problem in computer vision, with applications in mobile robotics and semantic scene understanding. In particular, mobile platform localization and guidance in urban environments can be enabled with accurate models of the various building facades in a scene. Toward that end, we present a system for detection, segmentation, and parameter estimation of building facades in stereo imagery. The proposed method incorporates multilevel appearance and disparity features in a binary discriminative model, and generates a set of candidate planes by sampling and clustering points from the image with Random Sample Consensus (RANSAC), using local normal estimates derived from Principal Component Analysis (PCA) to inform the planar models. These two models are incorporated into a two-layer Markov Random Field (MRF): an appearance- and disparity-based discriminative classifier at the mid-level, and a geometric model to segment the building pixels into facades at the high-level. By using object-specific stereo features, our discriminative classifier is able to achieve substantially higher accuracy than standard boosting or modeling with only appearance-based features. Furthermore, the results of our MRF classification indicate a strong improvement in accuracy for the binary building detection problem and the labeled planar surface models provide a good approximation to the ground truth planes.     

室内移动机器人的快速CCD视觉传感处理系统研究

针对传统方法直接对CCD采集的图像进行后续信号处理在实时性上的不足,特别是在高像素CCD构成的视觉系统中这种不足尤为突出。提出了一种快速CCD视觉传感处理方法,该方法包括动态灰度处理、Sobel后置优化、边缘跟踪Hough变化和传感器快速标定,能有效降低算法复杂性,提高整个视觉传感系统的实时性。在白行设计的ARM9（S3C2410）CCD传感的移动机器人平台上运行,实验证明具有较好的实时性。     

基于双目视觉移动机器人的路径规划和避障研究

提出了一种移动机器人路径规划和避障的系统设计方案,实现了移动机器人自主行进的路径规划和自动避障功能.详细说明了如何采用立体视觉实现对环境的探测,利用图像处理算法的组合分离出地面、背景、障碍物和目标物,采用边界不变矩实现障碍物和目标物的区分,改进了经典的人工势场法进行路径的规划,根据模糊控制原理设计了避障控制器和避障规则.实际的运行结果表明了该系统的可行性和有效性,该系统实现了移动机器人利用自身传感器感知环境信息,动态规划行进路径,成功躲避障碍物等功能.     

视觉传感系统的自主移动机器人路径规划方法

在未知环境下,机器人很难快速获取周边环境信息并建立实时环境地图,实现自主运行.为此提出基于视觉导航的方法,利用全景摄像机作为机器人的视觉传感器系统采集环境信息,将彩色地图进行HSI空间下模糊聚类图像分割,得到环境二值图像;将图像进行栅格化处理来构建环境地图,运用8方向连接的Dijkstra进行全局路径规划,计算出最优路径,从而实现移动机器人的快速、自主运动.经过仿真实验证明,该方法有效且可行.     

Centering behavior with a mobile robot using monocular foveated vision

A system for corridor following based on properties of the human visual system is presented. The robot extracts image features using an interest operator to compute sparse optical flow induced by the translatory motion of the robot. The available status information from the robot is used to compensate for the known rotatory movement of the image. Control of the robot is done by transforming the optical flow to ego-motion complex log space. The difference between the median flow extracted from the left and right peripheral visual areas is used to control the heading of the robot. Forward velocity is controlled by trying to keep the perceived optical flow constant.     

圆轨双链主动视觉机构标定体系的设计

针对圆轨双链主动视觉机构中摄像机动态标定问题,设计了一种在圆轨内侧安装均匀分布6组立体标靶块的标定体系.从保证摄像机在任意支链位置均能获得立体标靶块图像出发,分析了立体标靶块的合理布局和靶块尺寸,并给出了梯形靶面的编码标志点和非编码标志点的设计方案;基于小孔成像原理,建立了标定体系的视觉标定模型,实现了通过一块平面棋盘靶进行立体标靶块的非编码标志点空间坐标的测定,以及通过立体标靶块进行摄像机位姿的动态标定.实验结果表明了标定体系设计的合理性及用其动态标定的有效性.     

Giant magnetoresistance and its application in recording heads

A short overview is given of the giant magnetoresistance (GMR) effect and its device applications. As an example of present research topics, the optimization of the ferromagnetic-layer composition of spin valves for application in recording heads is discussed. The influence on output voltage and stability has been investigated in test structures and compared with material parameters that were experimentally obtained from unpatterned films. It is found that in practical applications a small addition of cobalt is advantageous, since, although the output voltage is decreased, it leads to an improvement in reproducibility, linearity, signal-to-noise ratio and dynamic range.     

Experiments on a floating underwater robot with a two-link manipulator

This article concerns experiments with a free-floating underwater robot with a two-dimensional, horizontal planar, two-link manipulator. Some dynamic models of underwater manipulators have been proposed, but only a few experiments have been carried out. Here, we derive a dynamic model for a free-floating underwater robot with a two-link manipulator, including the hydrodynamic forces, and validate the effectiveness of the model by simulation and experiment. We also show an experimental result using a resolved acceleration control method. These experimental results show the effectiveness the model and the control method. This work was presented in part at the 5th International Symposium on Artificial Life and Robotics, Oita, Japan, January 26–28, 2000     

Dual-coding representations for robot vision programming in Tekkotsu

We describe complementary iconic and symbolic representations for parsing the visual world. The iconic pixmap representation is operated on by an extensible set of “visual routines” (Ullman, 1984; Forbus et al., 2001). A symbolic representation, in terms of lines, ellipses, blobs, etc., is extracted from the iconic encoding, manipulated algebraically, and re-rendered iconically. The two representations are therefore duals, and iconic operations can be freely intermixed with symbolic ones. The dual-coding approach offers robot programmers a versatile collection of primitives from which to construct application-specific vision software. We describe some sample applications implemented on the Sony AIBO. David S. Touretzky is a Research Professor in the Computer Science Department and the Center for the Neural Basis of Cognition at Carnegie Mellon University. He earned his B.A. in Computer Science from Rutgers University in 1978, and his M.S. (1979) and Ph.D. (1984) in Computer Science from Carnegie Mellon. Dr. Touretzky’s research interests are in computational neuroscience, particularly representations of space in the rodent hippocampus and related structures, and high level primitives for robot programming. He is presently developing an undergraduate curriculum in cognitive robotics based on the Tekkotsu software framework described in this article. Neil S. Halelamien earned a B.S. in Computer Science and a B.S. in Cognitive Science at Carnegie Mellon University in 2004, and is currently pursuing his Ph.D. in the Computation & Neural Systems program at the California Institute of Technology. His research interests are in studying vision from both a computational and biological perspective. He is currently using transcranial magnetic stimulation to study visual representations and information processing in visual cortex. Ethan J. Tira-Thompson is a graduate student in the Robotics Institute at Carnegie Mellon University. He earned a B.S. in Computer Science and a B.S. in Human-Computer Interaction in 2002, and an M.S. in Robotics in 2004, at Carnegie Mellon. He is interested in a wide variety of computer science topics, including machine learning, computer vision, software architecture, and interface design. Ethan’s research has revolved around the creation of the Tekkotsu framework to enable the rapid development of robotics software and its use in education. He intends to specialize in mobile manipulation and motion planning for the completion of his degree. Jordan J. Wales is completing a Master of Studies in Theology at the University of Notre Dame. He earned a B.S. in Engineering (Swarthmore College, 2001), an M.Sc. in Cognitive Science (Edinburgh, UK, 2002), and a Postgraduate Diploma in Theology (Oxford, UK, 2003). After a year as a graduate research assistant in Computer Science at Carnegie Mellon, he entered the master’s program in Theology at Notre Dame and is now applying to doctoral programs. His research focus in early and medieval Christianity is accompanied by an interest in medieval and modern philosophies of mind and their connections with modern cognitive science. Kei Usui is a masters student in the Robotics Institute at Carnegie Mellon University. He earned his B.S. in Physics from Carnegie Mellon University in 2005. His research interests are reinforcement learning, legged locomotion, and cognitive science. He is presently working on algorithms for humanoid robots to maintain balance against unexpected external forces.     

Registering real and virtual imagery

Registering a virtual scene with a real scene captured by a video camera has a number of applications including visually guided robotic navigation, surveillance, military training and operation. The fundamental problem involves several challenging research issues including finding corresponding points between the virtual and the real scene and camera calibration. This paper presents our research in defining and mapping a set of reliable image features for registering the two imageries, extracting and selecting reliable control points for the construction of intrinsic and extrinsic camera parameters. A number of innovative algorithms are presented followed by extensive experimental analysis. An application of registering virtual database image with video image is presented. The algorithms we developed for calculating and matching linear structured features and selecting of reliable control points are applicable to image registration beyond virtual and real imageries.